[Mesa-dev] [PATCH] i965: allow image_size on float images
martin.peres at linux.intel.com
Fri Aug 21 09:09:30 PDT 2015
On 21/08/15 17:47, Martin Peres wrote:
> On 21/08/15 16:52, Francisco Jerez wrote:
>> The commit message should be prefixed "glsl:", this fix isn't
Ilia just warned me on IRC that I completely missed this comment...
I guess the top-posting threw me off (even if it was entirely
justified). This has not been my week it would seem! Good that is over!
>> Martin Peres <martin.peres at linux.intel.com> writes:
>>> This got missed because the piglit test only tested int images to
>>> avoid a
>>> combinatiorial explosion of format, targets, stages and sizes which
>>> takes more than 5 minutes to test on nvidia's driver.
>>> This patch also drops the IMAGE_FUNCTION_AVAIL_ATOMIC which is not
>>> to the image_size codepath but was not hurting in any way.
>> Well, it would have caused the intrinsic not to be exposed on unextended
>> GLES 3.1, which is definitely a bug. :P
> No, I had the same reaction at first, but I actually do not use
> I use _image_size_prototype which does not handle this flag.
>>> Signed-off-by: Martin Peres <martin.peres at linux.intel.com>
>>> src/glsl/builtin_functions.cpp | 3 ++-
>>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>> diff --git a/src/glsl/builtin_functions.cpp
>>> index 05dbe2e..0308aa6 100644
>>> --- a/src/glsl/builtin_functions.cpp
>>> +++ b/src/glsl/builtin_functions.cpp
>>> @@ -2695,7 +2695,8 @@ builtin_builder::add_image_functions(bool glsl)
>>> add_image_function(glsl ? "imageSize" :
>>> - &builtin_builder::_image_size_prototype, 1, atom_flags);
>>> + &builtin_builder::_image_size_prototype, 1,
>>> + flags |
>> Reviewed-by: Francisco Jerez <currojerez at riseup.net>
> Thanks, will push it now, before Emil pushes tags the first rc.
>>> ir_variable *
>>> mesa-dev mailing list
>>> mesa-dev at lists.freedesktop.org
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
More information about the mesa-dev